Journal of Computations & Modelling, vol.3, no.4, 2013, 11-24
ISSN: 1792-7625 (print), 1792-8850 (online)
Scienpress Ltd, 2013
Dimensioning of Electric Propulsion Motors for
War-Ships Using Finite Element Method
Angelos P. Moschoudis 1, Miltiades C. Filippou 2,
George J. Tsekouras 3 and Antonios G. Kladas 4
Abstract
The electrification of war-ship propulsion system is a very attractive solution, as it
provides increased safety, survivability, maneuverability, precise and smooth
speed control, reduced machinery space, low operation and maintenance costs,
low noise and low pollutant emission levels. In this paper the basic principles of
dimensioning for warships electric propulsion motors using finite element method
are presented. The respective example which will be shown synoptically is a low
rotational salient pole synchronous motor connected to the propeller axis directly
(without gearbox).
Keywords: Electric propulsion, finite element method, synchronous motor,
warship power system.
1
School of Electrical & Computer Engineering, National Technical University of Athens,
9 Iroon Polytechniou Street, 15780 Athens, Greece. E-mail: moshoudi@central.ntua.gr.
2
Mobile Communications Department, EURECOM, 06410 BIOT, France.
E-mail: filippou@eurecom.fr.
3
Hellenic Naval Academy, Terma Hatzikiriaku, Piraeus, GR-18539 Greece.
E-mail: tsekouras@snd.edu.gr.
4
School of Electrical & Computer Engineering, National Technical University of Athens,
9 Iroon Polytechniou Street, 15780 Athens, Greece. E-mail: kladasel@central.ntua.gr.
Article Info: Received : July 1, 2013. Revised : August 30, 2013.
Published online : December 1, 2013.
12
Dimensioning of Electric Propulsion Motors for War-Ships
1 Introduction to the Electrification of War-Ship Propulsion
System
Warships propulsion electrification offers significant anticipated benefits for
war ships in terms of reducing ship life cycle cost, increasing ship stealthiness,
payload, survivability and power available for non propulsion-uses, and taking
advantage of a strong electrical power technological and industrial base. Potential
disadvantages include higher near-term costs, increased technical risk, increased
system complexity, and less efficiency in full power operations ([1]).
In a ship with a mechanical-drive system, (see Fig. 1), the power-producing
capability of the ship's propulsion engines typically represents 75 percent to 85
percent of the ship's total power-producing capability. This power-producing
capability is devoted exclusively to ship propulsion and is not available for nonpropulsion uses, even when the ship is stationary or traveling at low speed.
Prime
Power
Conversion
and
Distribution
Generator
Main
Switch
Prime
Non
Propulsion
Electrical
Loads
Generator
Prime
Prime
Reduction
Gear
Generator
Prime
Prime
Reduction
Gear
Prime
Figure 1: Basic war ship mechanical drive system
Electric drive technology would change the way that war ships transmit
power from their engines to their propellers, as well as the way they manage and
Ang. Moschoudis, Milt. Filippou, G.J. Tsekouras and Ant. G. Kladas
13
distribute electrical power to both propulsion and non-propulsion systems. Ships
with an electric drive system can be designed so that a single set of engines
produces a common pool of electricity that is used for both ship propulsion and
the ship’s non-propulsion electrical loads (see Figure 2). Such a system is known
as an integrated electric-drive (IED) system or integrated power system (IPS).
IPS consists of the following set of modules: Power generation, propulsion
motor, power distribution, power conversion, power control, energy storage and
load. The above mentioned sets of modules and the IPS architecture combined
together provide the basis for designing, procuring, and supporting marine power
systems applicable over a broad range of commercial and war ship types ([2]).
Auxiliary Prime Movers
PM
Gen
PM
Gen
Power
Conversion
and
Distribution
Main Prime Movers
Non
Propulsion
Electrical
Loads
Switch
Motor
Drive
Motor
Switch
Motor
Drive
Motor
Main
Switch
Prime
Prime
Generator
Generator
Figure 2: Basic integrated electric drive system
In ships with integrated electric drive, the electricity produced by the
engines and generators is sent by cable to an electric switchboard that divides the
electricity into two flows one for the ship's propulsion needs, and one for the ship's
other electrical loads (see Figure 3). The switchboard can alter the distribution of
power between these two uses on a moment to moment basis, as needed, to meet
the ship's propulsion and non propulsion needs.
14
Dimensioning of Electric Propulsion Motors for War-Ships
Electric drive technology offers significant anticipated benefits in terms of
reducing ship life cycle cost, increasing ship stealthiness, payload, survivability,
and power available for non-propulsion uses, and taking advantage of a strong
electrical-power technological and industrial base ([3]). Depending on the kind of
ship in question and its operating profile (the amount of time that the ship spends
traveling at various speeds), a war ship with an integrated electric drive system
may consume 10 percent to 25 percent less fuel than a similar ship with a
mechanical drive system.
Electric propulsion makes possible the use of new propeller/stern
configurations, such as a podded propulsor, that can reduce ship fuel consumption
further due to their improved hydrodynamic efficiency, depending on the ship type
and the exact propeller/stern configuration used ([4]). Due to their design electric
drive systems can be highly automated and self monitoring. Consequently
reductions in maintenance and crew size would further reduce war ship life cycle
cost.
Electric propulsion promises to be significantly quieter acoustically than
mechanical drive. Since acoustic noise is an important component of a war ship's
overall detectability, ships equipped with electric drive promise to be less
detectible than ships equipped with mechanical drive technology. In a surface
combatant, electric drive's reduced fuel consumption can be translated into a
reduction in the amount of space aboard ship required for fuel storage. In addition,
by eliminating the need in a mechanical drive system, to install the engines,
reduction gears, shafts, and propellers in a long line running along the bottom of
the ship electric drive makes it possible to install the various parts of a surface
combatant's drive system in positions that may use space aboard ship in a more
efficient manner. In both these ways, electric drive may free up space that can be
used to carry additional payload (e.g., weapons or sensors). But electric drives
have also potential disadvantages in terms of higher near term costs, increased
program risk, increased system complexity, and less efficiency in full-power
Ang. Moschoudis, Milt. Filippou, G.J. Tsekouras and Ant. G. Kladas
15
operations. However, warships typically spend only a small fraction of their time
at full power.
Impedance
G
Main Generators
G
G
M
Filter
Electric
G
Medium Voltage Busbar
M
M
M
M
~
M
~
G
G
Auxiliary
Low Voltage Busbar 440 V
Low Voltage Loads
Emergency
Loads Back Up
Battery
Low Voltage
Generator
Busbar
G
Shore
Emergency
Low Voltage
Busbar
Low Voltage
Loads
Low Voltage Loads 115 V
Low Voltage Emergency
Loads Busbar 440 V
Low Voltage Emergency
Loads 440 V
Low Voltage Emergency
Loads 115 V
Figure 3: Analytical representation of a fully electrified war ship’s electric power
System
Much of the debate over electric drive concerns electric motors. The five basic
types in question synchronous motors, induction motors, permanent magnet
motors, superconducting synchronous motors, and superconducting homopolar
motors differ in terms of their technological maturity, power density, and potential
applicability to different war ships.
16
Dimensioning of Electric Propulsion Motors for War-Ships
The synchronous motor can be considered the most mature technologically in
application to warships. In this paper the basic principles of dimensioning for
warships synchronous propulsion motors using finite element method are
presented. Additionally, for this reason an under scale synchronous salient pole
motor design is analytically presented.
2 Introduction to Synchronous Propulsion Motor Design
In the electrification of war ship propulsion system considered, the
synchronous motor is the most mature technologically application. In such
applications low rotational speed salient pole synchronous motor connected to the
propeller shaft directly (without gear box) are used. Also the voltage and the
frequency of the electricity supply needed to be modified by a motor drive in order
the proper motor’s operation at the desired speed to be achieved. Here, a novel
two step design procedure for the design of a reliable multi-pole salient pole low
rotational speed synchronous propulsion motor is presented.
Engine
Generator
Switchboard
Motor
Drive
Motor
Electrical power for non
propulsion electrical loads
Figure 4: Block diagram representation of the war ship electric propulsion system
3 First-Step: Preliminary Design
The synchronous motor’s main dimensions are derived by using classical
formulae. These are calculated after the selection of motor’s parameters such as
Ang. Moschoudis, Milt. Filippou, G.J. Tsekouras and Ant. G. Kladas
17
nominal power, rotational speed, specific magnetic loading in the various parts,
and electric loading in the windings. Additionally the temperature rise and the
motor’s performance are calculated. The preliminary design is presented in the
flowchart of Figure 5.
Starting of Motors
Preliminary Design
ϕ, Nph , Stators Slots, τs, Iline ,
Sconductor Calculation
Construction of magnetization
curve
Motors Selection Design
Parameters ωr, B, q, KW , f, S
Calculation of Stators Slots
Teeth Core Depth Dimensions
Excitations full load ampereturns per pole calculation
Motors Poles Calculation,
Stators D L li, Pole Pitch τp
Vph , j Selection
Calculation of Copper, Stray,
Eddy Current and Stators Core
Power Losses
Air gap, Drotors, Rotors pole
characteristics Calculation
Design of field coil
Field winding and Stator
temperature rise calculation
Motors efficiency calculation
Figure 5: Flow chart used for synchronous motor’s main dimensions calculation
(preliminary design)
4 Second Step: Final Design – Numerical Simulation Using
Finite Element Modelling
The finite element model used in this application involves the solution of
some limiting cases of Maxwell’s partial differential equations. The magnetic
problems are those that can be considered as "low frequency problems", in which
displacement currents can be ignored. Displacement currents are typically relevant
to magnetics problems only at radio frequencies ([5]).
18
Dimensioning of Electric Propulsion Motors for War-Ships
Magnetostatic problems are problems in which the fields are time-invariant.
In this case, the field intensity H and flux density B must obey the following
equations:
∇×𝐻 =𝐽
(1)
∇∙𝐵 =0
(2)
𝐵 =𝜇∙𝐻
(3)
subject to a constitutive relationship between 𝐵 and 𝐻 for each material:
If a material is nonlinear (e.g. saturating iron, or alnico magnets), the permeability,
𝜇 is actually a function of 𝐵:
𝐵
𝜇 = 𝐻(𝐵)
(4)
Finite element method goes about finding a field that satisfies equations (1)(3) via a magnetic vector potential approach. Flux density is written in terms of
the vector potential, 𝐴, as:
𝐵 = ∇×𝐴
( 5)
Now this definition of 𝐵 always satisfies equation (2). Then equation, (1) can be
rewritten as ([5], [6]):
1
∇ × �𝜇(𝐵) ∇ × 𝐴� = 𝐽
(6)
Finite element method retains the form of equation (6), so that magnetostatic
problems with a non-linear B-H relationship can be solved. In the general 3D case
A is a vector with three components. However, in the 2 D planar and axisymmetric
cases, two of these three components are zero, leaving just the component in the
"out of the page" direction. The advantage of using the vector potential
formulation is that all the conditions to be satisfied have been combined to a single
equation.
The use of machines full geometry could have been applied in this problem
but this method is extremely time consuming due to the large number of nodes and
elements of the triangular mesh. In such cases, symmetry can be employed to
analyze only a fraction than the entire machine.
Ang. Moschoudis, Milt. Filippou, G.J. Tsekouras and Ant. G. Kladas
19
Generally, the number of rotor poles and stator slots can be divided by their
greatest common denominator in order the smallest region that can be analyzed to
be obtained. If the number of rotor poles to be analyzed is odd numbered, the ends
of the motor should be joined by anti-periodic (anti-symmetric) boundary
conditions. If the number of rotor poles were even periodic (symmetric) boundary
conditions would have been used instead. The design which will be presented has
32 salient rotor poles and a stator with 96 slots. Consequently the motor’s basic
structure consists of one pole and three stator slots. For the above mentioned
reasons the simulation of one motor’s pole has been adopted.
Once the motor’s basic structure has been determined the next step of the
design is performed with the application of the boundary conditions. In order to
improve the accuracy of the results the air-gap division in four parts has been
adopted.
Air-gap Limits
Rotor
Anti-periodic B5
Stator
Anti-periodic B4
Anti-periodic B10
Anti-periodic B6
Anti-periodic B7
Anti-periodic B8
Anti-periodic B9
Figure 6: Application of anti-periodic boundary conditions to the split motor in the
air gap
20
Dimensioning of Electric Propulsion Motors for War-Ships
Anti-periodic B9
Anti-periodic B4
Prescribed A
(Dirichlet)
Anti-periodic B3
Anti-periodic B1
Anti-periodic B2
Anti-periodic B4
Anti-periodic B1
Anti-periodic B2
Anti-periodic B9
Anti-periodic B3
Figure 7: Application of anti-periodic boundary conditions to the split motor in
rotor and stator
Dirichlet-type boundary condition A = 0, is defined on the boundary
(stators external diameter) in order magnetic flux to be kept from crossing the
boundary. The use of a finite element model will verify the preliminary results of
the classical design.
5 Application
The methodology presented has been applied to an under scale design of a 20
kVA, 120 rpm synchronous salient pole motor to be connected to a low rotational
speed propeller. The behavior of the machine has been investigated by considering
the use of the material M-43 for stators and rotors laminations, copper windings
with electric loading below 4 A/mm2, efficiency above 84 % respectively. At the
preliminary design stage the following main parameters have been calculated: an
air - gap width of 1.1 mm, number of stator slots 96, stator’s internal diameter
Ang. Moschoudis, Milt. Filippou, G.J. Tsekouras and Ant. G. Kladas
21
852 mm, stator’s external diameter 978 mm, rotor’s diameter 689,8 mm (without
salient poles), motors stator core length 109 mm, a single layer stator winding
with total slots area 464 mm2. Also all the above mentioned requirements were
calculated in order to be validated by using the method of finite elements.
The above mentioned machine configuration for each pole is shown in Figure
8 (during rotors rotation). Figure 9 gives the triangular mesh employed for the
finite element analysis of the machine involving approximately 6714 nodes and
12865 elements.
Figure 8: One pole machine design in the FEM model during rotors rotation
Figure 9: Mesh implemented in the FEM model
Figure 10 shows the magnetic field distribution corresponding to full load
conditions, while Figure 11 shows the respective flux density plot implemented in
the FEM model.
22
Dimensioning of Electric Propulsion Motors for War-Ships
Figure 10: Field distribution
Figure 11: Flux density plot implemented in the FEM model
The air gap’s arc flux density magnitude variations, the magnetic induction’s
harmonic components in the middle of the air-gap and the torque variations versus
the torque angle δ are shown in Figure 12, Figure 13 and Figure 14 respectively.
The requirements that were set in the preliminary design are validated (no
exceeding the specific magnetic loading of 1.5 T and the specific electric loading
of 4 A/mm2) and the resultant values of magnetization for the various generators
parts are 1.5 T for the stators teeth, 1.1 T for stators core and 1.34 T for rotors pole
respectively, while the specific electric loading of winding is not beyond 3.2
A/mm2.
1,0
0,5
0,0
0,00 0,02 0,04 0,06 0,08
Pole Pitch tp (m)
Figure 12: Flux density
magnitude variations
1,2
23
Torque (Nm) .
B (Wb/m^2)
1,5
0,9
0,6
0,3
2500
2000
1500
1000
500
0
0
1
3
5
7
9
11
13
15
17
Fourier coefficient.
Ang. Moschoudis, Milt. Filippou, G.J. Tsekouras and Ant. G. Kladas
0
Harmonic
Figure 13: Magnetic
induction’s harmonic
components
(air gap middle)
45
90 135 180
degrees
Figure 14 : Torque
variations versus angle δ
6 Conclusion
An integrated two step methodology for the classical design of a salient
pole, low r.p.m., synchronous motor for war ship electric propulsion application
has been presented. This method has been validated by numerical simulation
reducing the necessary number of finite element method executions and the
computational time during the selection of the generator’s dimensions through the
first step of classical design.
References
[1] R. O’ Rourke, Electric Drive Propulsion for U.S. Navy Ships : Background
and Issues for Congress, The Library of Congress, Congressional Research
Service, CRS Report for Congress, Order Code RL 30622, Security
Classification of Report, Unclassified, 31 July 2000, p. 3.
24
Dimensioning of Electric Propulsion Motors for War-Ships
[2] N. Doerry, Presentation of Designing All Electric Ships, 9th IMDC, (May 1619, 2006), 3-4.
[3] McCoy and J. Timothy, Powering the 21st Century Fleet, U.S. Naval Institute
Proceedings, (May 2000), 54-58.
[4] K. Bonner, Naval Propulsion for the 21st Century : The Azipod System, U.S.
Naval Institute Proceedings, (August, 1999), 74-76.
[5] D. Meeker, Finite Element Method Magnetics: User’s Manual, (October
2010), 7-8.
[6] J. Pyrhonen, T. Jokinen and V. Hrabovcova, Design of Rotating Electrical
Machines, J. Wiley & Sons, p. 6-10, 2008.